Research interests


Main research interests:

  • Phonology
  • Morphology
  • Tone (especially grammatical tone)
  • African languages
  • Creative adaptation of linguistic form
  • Documentary and descriptive linguistics

Current projects:

A reference grammar and tonal documentation of Seenku, an endangered Mande language of Burkina Faso (NSF DEL, BCS-1664355): Since 2012, I have been carrying out fieldwork on the southern dialect of Seenku (also known as Sembla, a Mande language of Burkina Faso) with the goal of writing a comprehensive reference grammar. I work with speakers both on site in Burkina Faso and with native speakers in New York City and Vienna. The language has a uniquely rich four-tone system, with numerous contour tones, providing a fertile ground for research on issues of tonal representation. In 2017, I received a three-year grant from NSF Documenting Endangered Languages to carry out this fieldwork, using innovative tools and methods to study tone, such as the xylophone surrogate language or automated tonal annotations (see below).

Seenku phonology in a xylophone surrogate language: Sembla music is largely performed on the balafon, a type of resonator xylophone common to West Africa. What makes this system fascinating from a linguistic standpoint is the heavy use of a surrogate language system, in which the spoken language Seenku is musically transposed onto the notes and rhythm of the xylophone. I am investigating the grammar of the surrogate language system in its two registers (“speaking” and “singing” styles), working with griot xylophonists in Burkina Faso as well as internationally-known Sembla musicians Mamadou Diabaté and his younger brother Seydou Diabaté.

Automated annotation of intermediate tonal representations: I am developing a computational tool to produce intermediate tonal annotation of texts and other recordings by extracting and normalizing F0 then categorizing the normalized tone into a number of levels, to be determined by the researcher. The output mimicks the common descriptive lingua franca seen in tonal descriptions, where a tonemic analysis such as a two-tone H/L system may be described in terms of levels 1-5 (numerically or represented with horizontal dashes) to capture effects like declination or downstep. This tool is intended both to aid in the annotation and eventual analysis of tone in language documentation as well as to test differences in the realization of tone in fluent speech vs. elicitation contexts. 

On the morphophonological mutability of phases (with Byron Ahn):  Drawing on our respective research on Dogon grammatical tone and Tongan definitive accent, we pursue the idea that the phonological content of phases (which results from morphophonology spelling out cycles of syntactic material) can be altered by morphophonological demands in later cycles. By virtue of having been spelled out, a computational unit (the complement to the phase head) loses all internal syntactic structure; no further changes can be made to the structure of that computational unit, since it has become essentially "a giant lexical compound" (Uriagereka 1999). This giant compound has phonological form (from passing through PF),  which we demonstrate can interact with morphology and phonology added at later cycles. Nevertheless, in many languages, changes to even the phonological material of a spelled-out computational unit are generally avoided, suggesting that this is a universally marked option. This has led some researchers to conclude that phonological changes to previously spelled out material are impossible. Instead, we account for these effects by proposing phase-phase faithfulness constraints (Faith-Phase), demanding that the output of one cycle of spell-out be preserved in the evaluation of the next cycle of spell-out. This preserves the finding that spelled out material ought not to change, but allows such changes to occur to satisfy morphophonological constraints (syntactic changes are still impossible), albeit not without penalty.

   © Laura McPherson 2017